1

Strength and Ductility Improvement of Recycled Aggregate Concrete 1

by Polyester FRP-PVC Tube Confinement 2

Chang Gao1, Liang Huang1*, Libo Yan2,3*, Ruoyu Jin4 and Bohumil Kasal2,3 3

1College of Civil Engineering, Hunan University, Changsha 410082, China 4 2Centre for Light and Environmentally-Friendly Structures, Fraunhofer Wilhelm-Klauditz-Institut WKI, 5

Bienroder Weg 54E, Braunschweig 38108, Germany 6 3Department of Organic and Wood-Based Construction Materials, Technical University of 7

Braunschweig, Hopfengarten 20, 38102 Braunschweig, Germany 8

4School of Environment and Technology, University of Brighton, Cockcroft Building 616, 9

Brighton, UK 10

*Corresponding authors: Liang Huang. Email: lianghuanghnu@gmail.com and Libo Yan. Email: 11 l.yan@tu-braunschweig.de 12

13 Abstract 14 In literature, studies on recycled aggregate concrete (RAC) with recycled aggregates (RAs) 15 originated from clay brick waste are rare, which is mainly attributed to the much lower 16 compressive strength of the RAC with recycled clay brick aggregates (RAC-RCBA) when 17 comparing with its normal aggregate concrete (NAC) counterpart. Nowadays it is well known 18 that fiber reinforced polymer (FRP) composites as lateral confining materials can improve the 19 strength and ductility of NAC significantly. In this study, FRP confining materials were used 20 to improve the compressive strength and ductility of the RAC-RCBA. Compared with 21 conventional synthetic glass or carbon FRP composites, polyester FRP (PFRP) and Polyvinyl 22 chloride (PVC) are much cheaper and show much larger tensile deformation capacity. 23 Therefore, this study investigated the axial compressive behavior of PFRP and PVC hybrid 24 tube encased RAC-RCBA (i.e., shortened as PFRP-PVC-RAC-RCBA) structure. This PFRP-25 PVC-RAC-RCBA system consisted of an RAC-RCBA core, encased by a PVC tube directly 26 and the PVC tube was further confined with a PFRP tube (i.e. PFRP tube-PVC-RAC-RCBA 27 specimen) or PFRP strips (i.e. PFRP strip-PVC-RAC-RCBA) at the outermost layer. Uniaxial 28 compression tests were performed on 33 PFRP-PVC-RAC-RCBA and 39 unconfined RAC-29 RCBA specimens to evaluate and compare the axial compressive behavior of PVC tube 30 encased RAC-RCBA, PFRP tube encased RAC-RCBA, PFRP tube-PVC-RAC-RCBA and 31 PFRP strip-PVC-RAC-RCBA columns. The tested variables included the number of PFRP 32 layers (3-, 6- and 9-layer), the type of PFRP confinement (in the configuration of tube or 33 strips) and the spacing of the PFRP strips (25 and 50 mm). The tested results demonstrated 34 that the PFRP-PVC hybrid confining system enhanced the compressive strength and axial and 35 lateral deformations of the RAC-RCBA pronouncedly, e.g. the increase in strength ranged 36 from 4.5% to 39.6%. The enhancement in strength and deformations was increased with a 37 thicker PFRP tube or strip. Both the PFRP tube-PVC-RAC-RCBA and PFRP strip-PVC-38 RAC-RCBA showed the similar axial compressive stress-stain behaviors. In addition, the 39 comparison of PFRP tube-PVC-RAC-RCBA with the glass/carbon FRP tube-RAC-RCBA 40 indicated that the GFRP and CFRP tube confinement resulted in much larger enhancement in 41 ultimate compressive strength of RAC-RCBA due to the much larger tensile modulus and 42 strength of these G/CFRP composites. However, PFRP-PVC tube confinement led to much 43 larger axial deformation of the RAC-RCBA compared with the G/CFRP tube confinement 44 due to the much larger tensile strain of the PFRP and PVC material. Furthermore, design-45 oriented compressive stress-strain models were developed for PFRP-PVC-RAC-RCBA 46 specimens. 47

Keywords: Recycled aggregate concrete (RAC); Recycled clay brick aggregates (RCBA); 48 Polyester fiber reinforced polymer (PFRP); PVC; Dual confinement; Compressive behavior 49

50

51

2

Nomenclature 52

RAC Recycled aggregate concrete d Diameter of cylindrical core concrete

RCBA Recycled clay brick aggregate fco Peak stress of unconfined RAC-

RCBA

RAs Recycled aggregates co Axial strain of unconfined RAC-

RCBA at peak stress

NAC Natural aggregate concrete fct

Peak stress of confined specimens

NAs Natural aggregates ct Lateral strain of confined specimens at

peak stress

FRP Fiber reinforced polymer l Axial strain of confined specimens at

peak stress

PFRP Polyester FRP fl Lateral confining pressure of FRP

GFRP Glass FRP fcu

Ultimate stress of the confined

specimens

CFRP Carbon FRP cu Ultimate axial strain of the confined

specimens at ultimate stress

AFRP Aramid FRP µ Ductility indices

BFRP Basalt FRP µt Dilation rate

PVC Poly Vinyl chloride fl’ Lateral confining pressure of the

composite confinement on concrete

core

RC Reinforced concrete flf Effective lateral confining pressure

provided by PFRP

AVG Average value flp Lateral confining pressure provided

by PVC

COV Coefficients of variation Epvc Elastic modulus of the PVC tube

SD Standard deviation ɛpvc Tensile strain in the hoop direction of

the PVC tube

D1 Diameter of core RAC tpvc Thickness of the PVC tube

D2 Diameter of PVC tube n Number of PFRP strips

H Height of concrete cylinders ke Effective confining coefficient of the

PFRP strip

Efrp Elastic modulus of PFRP tube Ae Effective confining area of the core

concrete

ɛfrp Tensile strain in the hoop direction of

PFRP tube

A Gross area of specimens including the

core concrete and external hybrid tube

tfrp Thickness of PFRP tube m Number of zone without confinement

among PFRP strip

tfrp’ Equivalent PFRP thickness for PFRP

strip-PVC-RAC-RCBA cylinders

s Spacing distance of the PFRP strip

bfrp Width of PFRP strip λ Related eigenvalue of ultimate stress

and the elastic modulus of confined

materials and core concrete

fl Lateral confining pressure Ec Elastic modulus of the RAC-RCBA

ffrp Tensile strength of FRP El Effective lateral confining stiffness of

the hybrid PFRP-PVC tube

tfrp Thickness of FRP

53

1 Introduction 54

The process of urbanization generated a large amount of construction and demolition waste 55 (CDW) which caused environmental pollution issues and difficulties to dispose those waste. 56

3

An abundant utilization of recycled aggregate concrete (RAC) could not only solve the 57 disposal issue of CDW but also reduce the consumption of natural resources [1-2]. RAC is an 58 environmentally friendly concrete in which part or all the natural aggregates (NAs) are 59 replaced by recycled aggregates (RAs) [2]. RAs are mostly sorted from crushed CDW. In 60 literature, most research of RAC focused on the use of RAs originating from old concrete 61 blocks. Indeed, except for the old concrete waste, clay brick waste also accounted for a large 62 portion of the CDW, i.e. up to 30-40% [3-4]. So, if recycled clay brick aggregates (RCBA) 63 originated from clay brick wastes can be used to produce RAC as structural concrete, this will 64 be a significant step for the development of sustainable concrete industry. However, RCBA 65 typically exhibited weaknesses when being used to produce RAC, i.e., high porosity and 66 variation in quality [5-9]. For example, because of the high porosity of RCBA, the crushing 67 index and water absorption of RAs can be significantly larger than those of NAs, i.e. the 68 crushing index and water absorption of RCBA might be 60% and 700% larger than those of 69 NAs, respectively. This can cause poor mechanical properties of the resulting RAC such as 70 the reduction in load carrying capacity and stiffness and increase in creep and shrinkage of the 71 RAC [5-7]. In addition, the complexity of RCBAs source results in the dispersion and 72 uncertainty of the mechanical properties of RAC in the aspects of flow ability, strength and 73 durability [8]. Thus, the mechanical properties of RAC-RCBA should be improved to expand 74 the range of their application considering the social and environmental benefits to use RCBA 75 [9]. 76

It has been widely accepted that confined concrete is an effective way to improve mechanical 77 properties of concrete [10-14]. Fiber reinforced polymer (FRP) composites such as glass FRP 78 (GFRP), carbon FRP (CFRP), and other FRP composites, i.e., basalt FRP (BFRP) [71] and 79 steel fibers [72], as one of the most effective confining materials for concrete, have been 80 widely used to improve the strength and ductility of natural aggregate concrete (NAC). 81 Concrete filled FRP tube (CFFT) is a hybrid structure that the pre-fabricated FRP tubes serve 82 as permanent formworks of fresh NAC and offer lateral confining pressure to enhance the 83 strength and ductility of the NAC core [13-19]. In literature, some recent studies have 84 investigated the behavior of RAC filled FRP tube [20-24]. For example, Gao et al. [21] 85 compared the compressive behavior of CFRP tube and GFRP tube encased RAC cylinders. 86 Ozbakkaloglu et al. [22] concluded that the RAC filled FRP tube with circular cross-sections 87 showed higher compressive strength when compared with the specimens with square cross-88 sections. Chen et al. [23] stated that the replacement ratio of RAs had a limited effect on the 89 CFRP confinement effectiveness for RAC. Choudhury et al. [24] concluded that the initial 90 stiffness of plain RAC-RCBA columns increased significantly with GFRP jacket confinement. 91 Ardavan et al. [25] even found that CFRP strengthening increased the load capacity of the 92 RAC beams and can be designed more load-affordable than the NAC beams. 93

However, synthetic GFRP and CFRP are expensive in their initial material price, non-94 degradable and non-recyclable. Against this background, recent researchers have used new 95 confining materials to gain environmental and economic benefits. Polyester fiber, as one 96 textile fiber, has advantages of large production, low price, degradability and appropriate 97 mechanical properties [26-31]. Therefore, polyester FRP (PFRP) has been used to confine 98 concrete columns. For example, Dai et al. [32] investigated the axial compressive behavior of 99 PFRP tube encased NAC and was compared with aramid FRP (AFRP) tube confined NAC. 100 This study showed that the PFRP confinement might not enhance the strength of the concrete 101 as much as that of the AFRP but PFRP improved the ultimate strain of the NAC more 102 pronouncedly and exhibited superior in cost performance. Ispir [33] stated that the PFRP 103 confinement improved the strength and ultimate axial strain of confined NAC and had high 104 deformation capacity which could be a good alternative in repairing or strengthening for 105 seismic-resistant RC structures. Saleem et al. [34-35] concluded that PFRP encased NAC 106 columns exhibited highly ductile behavior owing to the large rupture strain of the PFRP. 107 Pimanmas et al. [36] demonstrated that the stress-strain behavior of PFRP tube encased NAC 108 presented apparent softening stage and the compressive strength increased with more PFRP 109

4

thickness. Huang et al. [37] investigated the effect of PFRP confinement ratios and concrete 110 strength on compressive behavior of PFRP encased NAC. Huang et al. [38] also investigated 111 the effects of slenderness and size of on the PFRP tube confined concrete cylinders. 112

Poly Vinyl chloride (PVC) materials were also extensively utilized in the construction 113 industry as concrete moulds or pipes due to the high production, low price and stable service 114 performance [39-41]. The apparent merits of PVC materials include: (1) excellent corrosion 115 resistant, durability and mechanical stability, (2) smooth surface and good consistency with 116 other materials, e.g., concrete, water and acids, (3) large ultimate strain (i.e., ductile), (4) high 117 electrical insulation, (5) low diffusion for humidity, (6) low creep deformation, and (7) easy 118 for machining, cutting, gluing for fabrication versatility. Research [42] showed that PVC tube 119 did not lose strength significantly under the thermocycling tests. Nowack et al. [43] conducted 120 tests on PVC tubes that were buried under soil for 60 years. They found that the PVC tubes 121 did not deteriorate and was expected to serve for a further 50 years. Ranney et al. [44] found 122 that PVC tubes could adequately resist the influence of chloridion, salts, freeze thawing and 123 other chemical effects and maintain their long-term durability. Kurt [45] proposed PVC tube 124 encased NAC for structural application and concluded that the PVC confined NAC performed 125 like spiral steel confined concrete to afford effective ductility enhancement. Toutanji et al. 126 [46-48] conducted experiments on mechanical properties and durability of GFRP-PVC tube 127 encased NAC and found that the hybrid confinement enhanced the load carrying capacity and 128 ductility of concrete and provided excellent durability even under high corrosive conditions. 129 Wang et al. [49] stated that the thermal conductivity of PVC was only 0.45-0.6% of steel 130 which would afford more stable condition without apparent change of temperature for 131 concrete curing. Gupta et al. [50] investigated the PVC tube encased NAC under sea water for 132 six months. The tested results indicated that no degradation in the strength and ductility of the 133 confined NAC was observed and the PVC tube served as a safe jacket to protect the core 134 concrete. Gathimba et al. [51] demonstrated that the confinement effectiveness of PVC 135 confined NAC is dependent on the strength of concrete where the confinement ratio reduced 136 with a higher strength of the concrete. Jiang et al. [52] explored the influence of slenderness 137 ratio on CFRP-PVC tube encased NAC under uniaxial compression. 138

139 Based on the discussions above, it can be concluded that the cost-effective PFRP-PVC hybrid 140 tube confinement has the potential to improve the mechanical properties of RAC-RCBA. 141 Thus, in this study, PFRP-PVC tube encased RAC-RCBA (termed as PFRP-PVC-RAC-142 RCBA), for the first time, was proposed to improve the mechanical properties of RAC-RCBA. 143 This PFRP-PVC-RAC-RCBA system was consisted of a PFRP (at the outer layer of the tube) 144 and PVC (at the inner layer of the tube) hybrid tube and an RAC-RCBA infill as illustrated in 145 Fig.1 (a). In this system, the inner PVC tube not only serves as the permanent formwork for 146 the fresh RAC-RCBA but also serves as the permanent formwork for fabricating the outer 147 PFRP tube using the typical hand lay-up process [59]. It should be pointed out here that for 148 the typical hand lay-up process to make FRP tube, a mould (such as made of PVC or 149 aluminum tube) is needed and then the resin-impregnated fiber fabrics are wrapped onto the 150 mould. When the FRP tube is fully consolidated, it will be demoulded from the PVC or the 151 aluminum tube. In the case of PFRP-PVC hybrid tube situation, the demoulding process of 152 PFRP tube from the initial mould (e.g. a PVC tube) is not needed and in turn the construction 153 time can be further reduced. Specifically, the objectives of this study included: 154 1) To obtain the optimal design mix ratio for RAC-RCBA cylinder by testing different 155

replacement ratios of the RCBAs and different water-cement ratios; 156 2) To investigate the material properties of the PFRP and PVC composites with different 157

thicknesses by flat coupon tensile tests; 158 3) To investigate the axial compressive behavior of PFRP-PVC-RAC-RCBA cylinders 159

considering different column parameter effects: the number of PFRP layers (3, 6 and 9-160 layer), type of the PFRP confinement (i.e. in the configuration of PFRP tubes (see Fig1(b)) 161

5

and strips (see Fig 1(b)) and the spacing distance of the PFRP strips (25 mm and 55 mm); 162 and 163

4) To develop stress-strain modes for PFRP-PVC-RAC-RCBA in axial compression by the 164 regression analysis and iterative computations of the tested results. 165

166

a) PFRP tube-PVC-RAC-RCBA cylinder b) PFRP strip-PVC-RAC-RCBA cylinder

Fig 1 PFRP-PVC-RAC-RCBA schematic diagram

167

2 Experimental works 168 2.1 Material properties of RAC-RCBA 169 2.1.1 RAs 170 The RAs used in the experiments are shown in Fig.2, which were a mixture of RAs from 171 RCBAs and from recycled old concrete waste. The portion of the RCBAs accounted for 55% 172 of the RAs by mass. The properties of the RAs are listed in Table 1. The crushing index of the 173 RAs was 17.3% which meets the demands of crushing indexes of RAs specified in the 174 Chinese National Standard GB/T 25177-2010 [53]. The crushing index of the natural 175 aggregates was 10.7%. 176 177

Table 1 Physical property of RAs 178 Partial size

(mm)

Density

(kg/m3)

Porosity

(%)

Water absorption

(%)

Moisture content

(%)

Crushing index

(%)

5~10 1140 10 14.8 6.5 17.3

Fig.2 RAs used

2.1.2 Compression tests of RAC-RCBA 179

6

To evaluate the effects of replacement ratios of RAs and water-cement ratios on the 180 compressive strength of RAC-RCBA, two groups of experimental works were conducted: 181 Group A and Group B. The Group A consisted of three categories of RAC-RCBA cubes 182 (150×150×150 mm3) corresponding to three different replacement ratios of RCBA (i.e., 50%, 183 70% and 100%). For each category of the specimens, six RAC-RCBA cubic specimens were 184 constructed for the axial compression test. The details of the mix proportions of Group A are 185 listed in Table 2. The Group B consisted of three categories of RAC-RCBA cubes 186 (150×150×150 mm3) corresponding to three different water-cement ratios (i.e., 0.46, 0.50 187 and 0.56). For each category, six RAC-RCBA cubic specimens were also constructed for the 188 axial compressive test. The details of the mix proportions of Group B are listed in Table 3. In 189 Tables 2 and 3, r indicates the replacement ratio of the RAs for the natural aggregates and 190 𝜔 𝑐⁄ indicates the water-cement ratio. In this study, vibratory mixing technology was applied 191 to produce the RAC-RCBA by using a DT60 double-horizontal shafts mixer. The RAC-192 RCBA with vibratory mixing technology showed better in fluidity, load carrying capacity and 193 durability [56-57]. The vibratory mixing technology combined the vibratory function into the 194 traditional concrete stir to accelerate the stirring velocity, promoted the uniform distribution 195 of the aggregates and accelerated the hydration process of the cement to enhance the bond 196 between cement and aggregates. The crafts of vibratory mixing technology referred to a 197 secondary stirring: the whole cement, aggregates and half of the water were poured into the 198 mixer and rotated adequately for about 8-10s in the first step of stirring, then the left half of 199 the water was poured in and mixed for 30s for the second step of stirring. The fully mixed 200 RAC-RCBA was poured into the moulds and compacted by vibrator. After pouring, all 201 specimens were covered by soaking wet cloth and watered three times per day for 28 days. 202 203

Table 2 RAC-RCBA mix proportion of group A 204

No. 𝜔 𝑐⁄ Water

(kg/m3)

Cement

(kg/m3)

Fine aggregate

(kg/m3)

Coarse aggregate

(kg/m3)

RCBA

(kg/m3) r

1 0.40 237.5 600.9 520.2 0 1041.4 100%

2 0.40 237.5 600.9 520.2 312.4 729.0 70%

3 0.40 237.5 600.9 520.2 520.7 520.7 50%

205 Table 3 RAC-RCBA mix proportion of group B 206

No. 𝜔 𝑐⁄ Water

(kg/m3)

Cement

(kg/m3)

Fine aggregate

(kg/m3)

Coarse aggregate

(kg/m3)

RCBA

(kg/m3) r

1 0.46 292.3 635.4 571.0 310.1 731.0 70%

2 0.50 315.4 626.0 545.5 300.8 701.8 70%

3 0.56 362.1 650.0 545.5 301.0 702.1 70%

207 208 The cubic specimens at 7-day were tested under a DYE-2000 electro hydraulic compression 209 machine to determine the compressive strength. The tested results of the RAC-RCBA cubes 210 in Group A are illustrated in Fig. 3 and the coefficients of variation (COV) of the tested 211 compressive strength obtained from six cubic specimens are also presented to demonstrate the 212 degree of dispersion of tested results, which were calculated as the ratio of the standard 213 deviation and the mean value of the strengths. It shows that the compressive strength of the 214 RAC-RCBA with 70% replacement ratio of RAs was the largest. The tested strengths of the 215 RAC-RCBA specimens in the Group B are listed in Fig.4. It shows that for RAC with RAs 216 replacement ratio of 70%, the compressive strength showed a descending tendency with an 217 increase of the water-cement ratios from 0.40 to 0.50. The further increase of the water-218 cement ratio from 0.50 to 0.56 resulted in a slight increase in the strength of the RAC. The 219 slight increase in the strength for specimen with water/cement ratio of 0.56 might be 220 attributed to the following two-folds: (1) the high porosity and water absorption of the RCBA 221 limited the rates of hydration reaction of cement. With an increase of the water/cement ratio, 222 the RCBA was saturated with water which weakened the hydration reaction of the cement and 223

7

in turn increased the compressive strength slightly, (2) the old brick powder and old mortar 224 adhered at surface of the RCBA had a negative effect on the hydration reaction of the cement. 225 For specimen with a larger water/cement ratio of 0.56, the old powder and old mortar limited 226 the rates of hydration of the cement which resulted in a slight increase of the compressive 227 strength. In addition, the RAC-RCBA with water-cement ratio of 0.40 and RAs replacement 228 ratio of 70% had the highest compressive strength among all the categories in Group A and 229 Group B. 230 231

Fig.3 The influence of r on compressive

strength of RAC-RCBA

Fig.4 The influence of w/c on compressive

strength of RAC-RCBA

232 The failure modes of RAC-RCBA cubes are shown in Fig.5. In general, the experimental 233 observations indicated that all the RAC-RCBA specimens in Group A and Group B presented 234 a diagonal pyramid rupture face as illustrated in Fig 5(b), which was like that of the NAC 235 counterpart. Compared with the NAC, most of the RCBAs in the RAC were crushed and 236 broke under the compressive load, while most of the coarse NAs still maintained intact. This 237 phenomenon can be interpreted by the much lower crushing index of NAs compared with that 238 of the RAs. 239 240

a) Failure pattern of RAC-RCBA cube b) Rupture face of RAC-RCBA cube

Fig.5 Failure mode of RAC-RCBA cube 241 242

2.2 Preparation of PFRP-PVC-RAC-RCBA 243

2.2.1 Test matrix 244 To investigate the axial compressive behavior of PFRP-PVC-RAC-RCBA, 36 cylindrical 245 specimens (i.e. 3 unconfined RAC-RCBA and 33 confined RAC-RCBA cylinders) were 246 constructed and tested under axial compression. The tested variables included the number of 247 PFRP layer (i.e. 3, 6, and 9-layer), the type of the PFRP confinement (i.e. in the 248 configurations of tube (Fig. 1(a) and strips (Fig. 1(b))) and the spacing distance of the PFRP 249

40 50 60 70 80 90 100 11020

25

30

35

COV=0.012

COV=0.009

COV=0.010

Co

mp

ress

ive

stre

ng

th o

f R

AC

(M

Pa)

Replacement ratio of RAs %

W/C=0.40

0.35 0.40 0.45 0.50 0.55 0.6020

25

30

35

COV=0.013

COV=0.009

COV=0.011

COV=0.011

Com

pre

ssiv

e st

rength

of

RA

C (

MP

a)

Water cement ratio

70% RAs

8

strips (i.e. 25 mm and 50 mm). These specimens were classified into one category of 250 unconfined RAC-RCBA, one category of PVC tube encased RAC-RCBA (i.e. PVC-RAC-251 RCBA), one category of PFRP tube encased RAC-RCBA (i.e. PFRP-RAC-RCBA), and nine 252 categories of PFRP-PVC-RAC-RCBA (i.e. 3P-T, 3P-S25, 3P-S50, 6P-T, 6P-S25, 6P-S50, 9P-253 T, 9P-S25, and 9P-S50). For each category three identical specimen were tested. The details 254 of all the specimens are listed in Table 4. For the PFRP tube-PVC-RAC-RCBA specimens, 255 the letter P with a figure in the front was used to represent the specimens, i.e. 3P-T, 6P-T and 256 9P-T indicates 3-layer, 6-layer and 9-layer PFRP tube-PVC-RAC-RCBA specimens, 257 respectively. For PFRP strip-PVC-RAC-RCBA specimens, the figure in front of the letter P 258 denotes the number of layers of the PFRP strip and the figure behind the letter S denotes the 259 spacing distances of the PFRP strip, e.g., the specimen code 3P-S25 denotes PFRP strip-PVC-260 RAC-RCBA specimen with 3 layers of PFRP strips and spacing distance of 25 mm. The 261 diameter of the core RAC D1 and the height of core RAC are 152 mm and 305 mm, 262 respectively. The external diameter of the PVC tube D2 and the thickness of PVC tube is 160 263 mm and 4 mm, respectively. 264 265

Table 4 Characteristics of tested cylindrical specimens 266

No. Specimen D1 of core RAC

(mm)

D2 of PVC tube

(mm)

Height

(mm)

PFRP

layer

PFRP confined

modes

1 RAC-RCBA 152 — 305 — —

2 PVC-RAC-

RCBA 152 160 305 — —

3 PFRP-RAC-

RCBA 152 — 305 6 Tube

4 3P-T 152 160 305 3 Tube

5 3P-S25 152 160 305 3 spacing distance

25mm

6 3P-S50 152 160 305 3 spacing distance

50mm

7 6P-T 152 160 305 6 Tube

8 6P-S25 152 160 305 6 spacing distance

25mm

9 6P-S50 152 160 305 6 spacing distance

50mm

10 9PT 152 160 305 9 Tube

11 9P-S25 152 160 305 9 spacing distance

25mm

12 9P-S50 152 160 305 9 spacing distance

50mm

267 For all the confined RAC-RCBA, the concrete mix proportion of the RAC-RCBA followed 268 the specimen with the highest compressive strength given in Section 2.1, namely, the RA 269 replacement ratio of 70% and the water-cement ratio of 0.40. The tested compressive strength 270 of the RAC-RCBA at 28-day based on six identical specimens was 30.6 MPa, and the 271 standard deviation was 0.45 MPa. 272 273 2.2.2 PFRP materials 274 The bidirectional polyester textile with an areal density of 250 g/m2 and a thickness of 0.55 275 mm were used to fabricate the PFRP tubes. Fig.6 (a) shows the polyester textile used for the 276 study. The textile was made by continuous polyester fibre filaments and these long fibres 277 were oriented in the wrap and weft directions of the textile with an angle of 90o. According to 278 ASTM D3039-M08 [54], flat coupon tensile tests were conducted on PFRP laminate to obtain 279 the tensile strength, strain and modulus. The details of the PFRP laminates used for the flat-280 coupon tests are shown in Fig. 6. Aluminum bars of 50 mm in length and 25 mm in width 281 were glued to the both ends of the PFRP laminates to avoid premature failure. The PFRP 282

9

laminates with 1, 3, 6 and 9 layers of polyester fabric were tested under the tensile load by 283 using the MTS CMT4204 universal testing machine. 284 285

a)

b) Dimension of the PFRP flat coupon

Fig 6 (a) The bidirectional polyester textile and (b) Dimension of the PFRP flat coupon

286 2.2.3 PVC materials 287 The 4 mm-thick PVC tubes were used for the PVC-RAC-RCBA, PFRP tube-PVC-RAC-288 RCBA and PFRP strip-PVC-RAC-RCBA specimens. Tensile tests were also carried out on 289 the PVC flat coupon to determine their tensile properties. The size of the PVC coupon used 290 for the tensile test is shown in Fig.7 according to the GB/T 8804.2-2003 [55], where 291 A=115mm, B=25mm, C=33mm, D=6mm, E=14mm, F=25mm, G=25mm, H=80mm and 292 I=4mm, respectively. 293

Fig. 7 The dimension of the tested PVC coupon

294 2.2.4 Tensile behavior of PFRP, PVC and comparison with synthetic C/GFRP 295

PFRP COUPON AluminumAluminum

50mm 50mm

250mm

25m

m2m

m

t

A

H

C

G

B DI

10

In this study, the tensile properties of PFRP and PVC were compared with the synthetic 296 GFRP and CFRP composites which used for GFRP and CFRP tube confined RAC-RCBA 297 [21]. The tensile stress-strain curves of PFRP, PVC and GFRP and CFRP obtained from flat-298 coupon tensile tests are shown in Fig. 8, 9 and 10. The PFRP composites showed a nonlinear 299 response until the peak stress and then the specimens failed suddenly, which was independent 300 of the thickness of the PFRP laminate. The PVC showed an initial linear response to the level 301 of the peak stress and followed by a stress plateau with significant enhancement in the tensile 302 strain, indicating a ductile behavior. For both the GFRP and CFRP composites, their curves 303 exhibited an approximate elastic response until the peak stress. For both the GFRP and CFRP 304 composites, their tensile stress-strain curves exhibited an approximate elastic response until 305 the peak stress. The damage of the CFRP and GFRP composites in tension was a progressive 306 and brittle failure process. More details on the progressive degradation failure process of FRP 307 composites and how to model the progress numerically were introduced by Riccio et al. [68-308 70]. The tensile properties of PFRP, PVC, GFRP and CFRP composites are listed in Table 5. 309 The tensile strength, ultimate strain and elastic modulus were the mean value of the tested 310 results from the flat-coupon tensile tests and the standard deviations are given to present the 311 dispersion of the tested results. The tensile modulus and tensile strength of GFRP and CFRP 312 were significantly larger than those of PFRP and the PVC. However, the tensile strain at 313 break of PVC and PFRP were much larger than those of the CFRP and GFRP, e.g. the tensile 314 strain of PVC was 10 times that of the GFRP. In addition, an increase in the thickness of the 315 PFRP composite resulted in an increase in tensile strength and modulus, but a slight reduction 316 in the tensile strain. The polymer resin used for PFRP, CFRP and GFRP were the same, i.e. 317 epoxy matrix with a commercial name of JN-C3P adhesive obtained from GOODBOND 318 construction technic co., Ltd in China. The tensile strength, compressive strength, flexural 319 strength and tensile elastic modulus of the JN-C3P resin provided by the supplier are 55 MPa, 320 83.6 MPa, 75 MPa and 3.5 GPa, respectively. 321 322

Fig 8 Tensile stress-strain behavior of PFRP Fig. 9 Tensile stress-strain behavior of PVC

0.00 0.01 0.02 0.03 0.04 0.050

10

20

30

40

50

Str

ess

(MP

a)

Strain

1-ply

3-ply

6-ply

9-ply

0.00 0.04 0.08 0.12 0.16 0.200

10

20

30

40

50

Str

ess

(MP

a)

Strain

PVC

11

Fig. 10 Tensile stress-strain behavior of CFRP and GFRP [21]

323 Table 5 Tensile properties and standard deviation (SD) of PFRP, PVC, GFRP and CFRP composites 324

Group

Thickness of

FRP (mm)

Tensile

strength

(MPa)

SD

(MPa)

Ultimate

strain (%)

SD

(MPa)

Elastic

Modulus

(GPa)

SD

(MPa)

PFRP1 0.66 27.1 1.5 4.4 0.4 0.9 0.08

PFRP3 1.72 31.1 1.3 4.1 0.2 1.2 0.10

PFRP6 3.18 41.7 1.7 3.8 0.1 1.6 0.12

PFRP9 5.12 45.1 1.4 3.4 0.2 2.0 0.13

PVC 4.00 22.5 0.7 16.0 0.1 0.3 0.03

GFRP 2.62 967.0 - 1.6 - 60.8 -

CFRP 0.96 3200.0 - 1.5 - 213.0 -

325 2.3 Fabrication of confined specimens 326 The PVC tube and the polyester fabrics were firstly cut into designated size based on the 327 dimension of the specimens listed in Table 4. The PFRP tubes and PFRP-PVC tubes were 328 produced by a hand lay-up process as illustrated in Fig. 11[38]. The fabrication of the PFRP 329 tube by hand lay-up process mainly included the following steps: 1). Cut polyester textile to 330 designated size based on the diameter of the polymer FRP tube used, 2) surface preparation of 331 hollow PVC tube mould with a thin release plastic film for easy demould, 3) preparation of 332 epoxy mixture, 4) wrapped epoxy-impregnated polyester fabric to the PVC mould, 4) curing 333 of the PFRP tubes, and 5) demoulding PFRP tube from the PVC mould. For the PFRP-PVC 334 tube with PFRP tube or PFRP strips, the process was similar. The PVC tube was used as the 335 mould directly and then the epoxy-impregnated polyester fabrics were wrapped into the PVC 336 tube to make the PFRP-PVC tube. For the fabrication of PFRP strip-PVC tube, the position of 337 polyester fabric strip was initially oriented and labelled on the PVC tube. The air bubble and 338 additional epoxy resin were squeezed out when the PFRP sheets were rolled onto the PVC 339 tube. For all the confined RAC-RCBA specimens, vibratory mixing technology was used as 340 described in Section 2.1.2. To avoid the premature failure of the confined specimens, two 341 narrow PFRP strips were wrapped on both ends of the PFRP-RAC-RCBA, PVC-RAC-RCBA 342 and the PFRP -PVC-RAC-RCBA specimens. 343

a) PFRP strip-PVC tube b) PFRP tube c) All the confined specimens

Fig.11 Fabrication of the specimens

0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.0160

500

1000

1500

2000

2500

3000

3500

4000

Str

ess

(MP

a)

Strain

CFRP

GFRP

12

344 2.4 Test instrumentation and procedures 345 The MTS SANS YAW6506 electro-hydraulic testing machine was used for the uni-axial 346 compression tests (Fig. 12(a)). The loading process was executed by a displacement-control. 347 As illustrated in Fig. 12(b)-(c), four axial strain gauges (i.e., SG1, SG2, SG3, SG4) and four 348 lateral strain gauges (i.e., SG5, SG6, SG7, SG8) were installed on the mid-height of the 349 cylinders for measurement of axial strain and hoop strain, and two extra axial strain gauges 350 were installed on each end of PFRP tube-PVC-RAC-RCBA cylinders (i.e., SG9 and SG11 at 351 the top of PFRP tube-PVC-RAC-RCBA specimens, SG10 and SG12 at the bottom of PFRP 352 tube-PVC-RAC-RCBA specimens), respectively. The applied load and vertical deformation 353 were recorded by MTS system of the compression testing machine. The axial strain, lateral 354 strain, applied load and vertical deformation were measured and recorded simultaneously. 355 356

a) MTS pressure testing machine b) Placement of strain gauges

c) Elevation of strain gauges on specimens

Fig.12 Test instrumentation

3 Results and discussion 357 3.1 Failure modes 358 For the unconfined RAC-RCBA cylinders, the cracks emerged around the surface of the 359 cylinders after the applied load was up to 40% of its peak strength. The load decreased rapidly 360 after reached its peak strength fco and the cylinders finally failed in a brittle manner with 361 several major vertical cracks (as illustrated in Fig 13(a)). For PVC-RAC-RCBA (i.e. only 362 with PVC tube confinement) specimens, the specimens failed with several longitudinal cracks 363 at the PVC tube and the tube were bulged apparently as illustrated in Fig. 13(b). The obvious 364 bulge of the PVC tube might be caused by the significant lateral expansion of the inner RAC-365

B B

Load

Load

PFRP COUPON AluminumAluminum

50mm 50mm

250mm

25

mm

2m

m

t

PFRP

RAC core

PVC

Overlapping lengthSG1

SG3

SG5 (9, 10)

SG7 (11,12)

SG4

SG8 SG2SG6

h/6

h/6

h/3

h/3

h/6

h/6

h/3

h/3

h/6

h/6

h/3

h/3

h/2

h/2

h/2

h/2

h

PFRP tube-PVC (PFRP)

confined RAC

25mm PFRP

strip-PVC

confined RAC

50mm PFRP

strip-PVC

confined RAC

PVC confined

RACuconfined RAC

h/2

h/2

h/2

h/2

PFRP+PVC环箍间距25mm试件

PFRP+PVC环箍间距50mm试件

SG5SG1 SG5SG1 SG5SG1 SG5SG1 SG5SG1

SG10

SG9

13

RCBA core due to the confinement provided by the PVC tube, which possessed a 366 significantly large tensile strain (Table 5). For the PFRP-RAC-RCBA, only a single 367 longitudinal crack appeared along the PFRP tube, no obvious bulging of the tube was 368 observed, as shown in Fig.13(c). This phenomenon in the different level of bulge of the PFRP 369 and PVC tube might be interpreted by the less ultimate tensile strain of the PFRP. For PFRP 370 tube-PVC-RAC-RCBA specimens with different thicknesses of the PFRP tube, their failure 371 modes were similar and had a single longitudinal crack in the PFRP-PVC tubes. No 372 debonding between PFRP and PVC tubes was observed. In addition, apparent bulging of the 373 PFRP-PVC tubes was observed, as illustrated in Fig. 13(d), (e) and (f). Thus, the increase in 374 the PFRP thickness did not change the general failure mode of the PFRP-PVC-RAC-RCBA. 375 For PFRP strip-PVC-RAC-RCBA specimens, the failure modes were different from the PFRP 376 tube-PVC-RAC-RCBA. The PFRP-strip-PVC-RAC-RCBA specimens had circumferential 377 cracks at the PVC tubes and the longitudinal cracks did not go through the whole height of the 378 PFPR-strip-PVC tubes, as shown from Fig 13(g)-(l). The appearance of the hoop rupture at 379 the PVC tubes attributed to the local strengthening effect of the PFRP strips on the PVC tubes. 380 As can be further observed, the increase of the thickness of the PFRP strips and the change of 381 the spacing of the PFRP strips did not change the failure modes of the PFRP strip-PVC-RAC-382 RCBA. 383

a) RAC-

RCBA

b) PVC-RAC-

RCBA

c) PFRP-RAC-

RCBA d) 3P-T e) 6P-T f) 9P-T

g) 3P-S25 h) 3P-S50 i) 6P-S25 j) 6P-S50 k) 9P-S25 l) 9P-S50

Fig.13 Failure modes of different confined RAC-RCBA specimens

384 3.2 Axial stress-strain behavior 385

3.2.1 The typical compressive stress-strain curves of PFRP-PVC-RAC-RCBA 386

The typical compressive stress-strain curves of unconfined RAC-RCBA, PVC-RAC-RCBA, 387 6-layer PFRP-RAC-RCBA, 6-layer PFRP tube-PVC-RAC-RCBA, 6-layer 25 mm and 50 mm 388 PFRP strip-PVC-RAC-RCBA specimens are illustrated in Fig 14. One representative 389 compressive stress-strain curve obtained from one of the three specimens for each type of 390 FRP confined RAC-RCBA specimens was used to plot the figure. For PVC-RAC-RCBA and 391 PFRP-RAC-RCBA specimens, their curves can be characterized by two distinct stages: the 392 first linear elastic response to the peak stress and followed by a nonlinear descending stage. 393 No obvious transition zone between the first linear stage and the second non-linear stages can 394 be found. In general, the stress-strain curve of PFRP-PVC-RAC-RCBA can be divided into 395 three zones: the first linear elastic ascending stage close to the peak stress, the second distinct 396 short non-linear ascending zone until the peak stress and the third non-linear descending stage 397 after the peak stress. Fig 15 gives a schematic view of the compressive stress-axial strain 398 curve for PFRP-PVC-RAC-RCBA specimens, which can be represented by two key points: 399

14

the transitional point (TP) corresponds to the peak stress point (i.e., peak stress fct and 400 corresponding peak strain εct point) and the ultimate point (UP) corresponds to the end of the 401 curves at the ultimate state (i.e., ultimate strain εcu and corresponding ultimate stress fcu). 402 Unlike the traditional CFRP and GFRP confined RAC (e.g. when the CFRP and GFRP had a 403 thickness of 4 and 6 layers in Ref. [21], where 4-layer and 6-layer GFRP-RAC and CFRP-404 RAC showed an ascending branch in the second stage of the compressive stress-strain curves) 405 which showed a typical bilinear response with a linear and ascending stage after the transition 406 zone, all the PFRP-RAC-RCBA, PVC-RAC-RCBA, and PFRP-PVC-RAC-RCBA showed 407 the descending stages at the second stage. In the other words, the ultimate strain does not 408 happen at the peak stress point but develops until the failure of specimens due to the ductility 409 of materials [38]. Therefore, the PFRP-RAC-RCBA, PVC-RAC-RCBA and PFRP-PVC-410 RAC-RCBA specimens can be defined as weakly-confined specimens, although enhancement 411 in compressive strength and ductility were obtained. 412

413

Fig.14 Stress-strain curves of different confinement modes

Fig.15 Typical axial compressive stress-strain curves of PFRP tube-PVC-RAC-RCBA and PFRP strip-PVC-RAC-

RCBA

3.2.2 The influence of confinement types on compressive stress-strain behavior 414

0

5

10

15

20

25

30

35

40

45

50

-0.02 -0.01 0 0.01 0.02 0.03Axial strainLateral strain

Str

ess

(MP

a)

6-layer and 50 mm PFRP strip-PVC-RAC-RCBA

6-layer and 25 mm PFRP strip-PVC-RAC-RCBA

6-layer PFRP tube-PVC-RAC-RCBA

6-layer PFRP-RAC-RCBA

4 mm PVC-RAC-RCBA

Unconfined RAC-RCBA

ct

fct

cu

fcu

UP

Ax

ial

stre

ss (

MP

a)

Axial strain

PFRP-PVC-RACTP

15

The overall patterns of the stress-strain behaviors of PFRP-RAC-RCBA, 6-layer PFRP tube-415 PVC-RAC-RCBA and 6-layer PFRP strip-PVC-RAC-RCBA with spacing distances of 25 416 and 50 mm illustrated in Fig. 14 are similar, which consisted of three stages: the initial elastic 417 linear ascending stage, the second nonlinear ascending stage and the final descending stage. 418 The PVC-RAC-RCBA specimens showed steeper descending stage and less peak stress than 419 PFRP-RAC-RCBA, PFRP strip-PVC-RAC-RCBA and PFRP tube-PVC-RAC-RCBA 420 specimens, but less slope of descending stage than that of the plain RAC-RCBA. The stress-421 strain curve of 6-layer PFPR-RAC-RCBA exhibited a sharper turning point from the peak 422 stress point to the descending stage and larger slope of the descending stage than those of 6-423 layer PFRP tube-PVC-RAC-RCBA specimens. While the ultimate strains at UP point of 6-424 layer PFPR-RAC-RCBA were like those of 6-layer PFRP tube-PVC-RAC-RCBA but the 425 peak stress at TP point of 6-layer PFPR-RAC-RCBA was lower than that of 6-layer PFRP 426 tube-PVC-RAC-RCBA and was close to that of 6-layer and 25 mm spacing distance PFRP 427 strip-PVC-RAC-RCBA. The stress-strain curves of 6-layer PFRP strip-PVC-RAC-RCBA 428 presented obviously more steep descending stage than that of 6-layer PFRP tube-PVC-RAC-429 RCBA, and the ultimate axial strain and peak stress of 6-layer PFRP strip-PVC-RAC-RCBA 430 were less than those of 6-layer PFRP tube-PVC-RAC-RCBA. 431

3.2.3 The influence of PFRP thickness on stress-strain behavior 432 According to Fig. 16(a), the stress-strain behavior of PFRP tube-PVC-RAC-RCBA showed a 433 little steeper trend at the second ascending stages with a decrease of PFRP thickness and more 434 placid descending stages with an increase of PFRP thickness, and the TP points performed 435 gentler transition with an increase of PFRP thickness. The stress-strain curves of PFRP strip-436 PVC-RAC-RCBA of the same spacing distance of PFRP strip in Fig. 16 (b) and (c) exhibited 437 similar trend to those of PFRP tube-PVC-RAC-RCBA that the initial ascending stages were 438 coincided with that of plain RAC-RCBA, the slopes of the second ascending stages became 439 lower and the descending stages turn more placid with an increase of PFRP thickness. The 440 ultimate strains at UP point and peak stress at TP point were higher for all PFRP-PVC-RAC-441 RCBA specimens with an increase of PFRP thickness. 442

a) PFRP tube

0

5

10

15

20

25

30

35

40

45

50

-0.02 -0.01 0.00 0.01 0.02 0.03Axial strainLateral strain

Str

ess

(MP

a)

9-layer PFRP tube-PVC-RAC-RCBA

6-layer PFRP tube-PVC-RAC-RCBA

3-layer PFRP tube-PVC-RAC-RCBA

6-layer PFRP tube-RAC-RCBA

4 mm PVC-RAC-RCBA

Unconfined RAC-RCBA

16

b) PFRP strip with spacing of 25 mm

c) PFRP strip with spacing of 50 mm

Fig.16 Stress-strain curves of different PFRP layers-PVC confined RAC-RCBA cylinders

443

3.2.4 The influence of spacing distance of PFRP strip on stress-strain behavior 444 The stress-strain curves in Fig. 17 present similar trend with different spacing distances of 445 PFRP strip to the typical stress-strain behavior in Fig. 15. The ascending stages of 3-layer and 446 6-layer PFRP strip-PVC-RAC-RCBA groups exhibited the similar slope with different 447 spacing distance of PFRP strip while the slopes of the descending stages became lower with a 448 decrease of spacing distance of PFRP strip. For 9-layer PFRP strip-PVC-RAC-RCBA group, 449 the slopes of the second ascending stages increased obviously with a decrease of spacing 450 distance of PFRP strip, besides the descending stages showed more placid with a decrease of 451 spacing distance of PFRP strip. For all PFRP strip-PVC-RAC-RCBA specimens, the ultimate 452 strains at UP point and peak stress at TP point increased with a decrease of spacing distance 453 of PFRP strip. 454

0

5

10

15

20

25

30

35

40

45

50

-0.02 -0.01 0.00 0.01 0.02 0.03

9-layer PFRP strip-PVC-RAC-RCBA

6-layer PFRP strip-PVC-RAC-RCBA

3-layer PFRP strip-PVC-RAC-RCBA

Unconfined RAC-RCBA

Lateral strain Axial strain

Str

ess

(MP

a)

0

5

10

15

20

25

30

35

40

45

50

-0.02 -0.01 0.00 0.01 0.02 0.03

9-layer PFRP strip-PVC-RAC-RCBA

6-layer PFRP strip-PVC-RAC-RCBA

3-layer PFRP strip-PVC-RAC-RCBA

Unconfined-RAC-RCBA

Lateral strain Axial strain

Str

ess(

MP

a)

17

a) 3-layer PFRP

b) 6-layer PFRP

c) 9-layer PFRP

Fig.17 Compressive stress-strain curves of different spacing distance of PFRP strip-PVC confined RAC-RCBA

cylinders 455 3.3 Discussion on tested results 456 3.3.1 Load carrying capacity and ductility analysis 457

0

5

10

15

20

25

30

35

40

45

50

-0.02 -0.01 0.00 0.01 0.02 0.03Lateral strain Axial strain

Str

ess

(MP

a)

3-layer PFRP tube-PVC-RAC-RCBA

3-layer and 25 mm PFRP strip-PVC-RAC-RCBA

3-layer and 50 mm PFRP strip-PVC-RAC-RCBA

Unconfined RAC-RCBA

0

5

10

15

20

25

30

35

40

45

50

-0.02 -0.01 0.00 0.01 0.02 0.03Lateral strain Axial strain

Str

ess

(MP

a)

6-layer PFRP tube-PVC-RAC-RCBA

6-layer and 25 mm PFRP strip-PVC-RAC-RCBA

6-layer and 50 mm PFRP strip-PVC-RAC-RCBA

Unconfined RAC-RCBA

0

5

10

15

20

25

30

35

40

45

50

-0.02 -0.01 0.00 0.01 0.02 0.03

9-layer PFPR tube-PVC-RAC-RCBA

9-layer and 25 mm PFPR strip-PVC-RAC-RCBA

9-layer and 50 mm PFPR strip-PVC-RAC-RCBA

Unconfined RAC-RCBA

Str

ess

(MP

a)

Lateral strain Axial strain

18

All the tested results are listed in Table 6, where fco and co are the peak stress and 458

corresponding axial strain of unconfined RAC-RCBA, respectively, fct and l , ct are the 459

peak stress, corresponding lateral and axial strain of confined specimens, fcu and cu are the 460

ultimate stress and corresponding ultimate axial strain of the confined specimens. The lateral 461 confining pressure fl is calculated as Eq. (1) [21], where the ffrp and tfrp are the tensile strength 462 and thickness of the FRP or PVC, respectively, d is the diameter of the concrete cylindrical 463 specimens. 464

𝑓𝑙 = 2𝑓𝑓𝑟𝑝𝑡𝑓𝑟𝑝/𝑑 (1) 465

Table 6 Tested results 466

Specimen fco

(MPa) co

(10-2)

fct

(MPa) ct

(10-2)

l

(10-2)

fl

(MPa) fcu

(MPa) cu

(10-2)

RAC-RCBA 30.6 0.2 — — — — — —

PVC-RAC-RCBA 30.6 0.2 30.6 0.21 0.02 1.20 6.4 0.79

PFRP-RAC-RCBA 30.6 0.2 38.4 0.30 0.14 2.31 28.2 2.55

3P-T 30.6 0.2 37.1 0.36 0.18 2.32 24.5 1.64

3P-S25 30.6 0.2 33.1 0.26 0.16 1.64 20.8 1.25

3P-S50 30.6 0.2 31.9 0.21 0.12 1.44 13.3 1.00

6P-T 30.6 0.2 40.9 0.45 0.28 3.73 28.3 2.54

6P-S25 30.6 0.2 39.3 0.34 0.21 2.21 20.9 1.75

6P-S50 30.6 0.2 36.0 0.29 0.08 1.78 20.4 1.20

9P-T 30.5 0.2 42.6 0.50 0.32 5.17 34.7 2.84

9P-S25 30.5 0.2 39.8 0.41 0.40 2.79 31.7 2.02

9P-S50 30.5 0.2 36.5 0.38 0.30 2.12 28.4 1.59

467 According to the tested results, the increments in the compressive strength enhancement at TP 468 point in Fig. 15 by the different confinements can be calculated. The corresponding increase 469 ratios of different confined RAC-RCBA referring to the unconfined RAC-RCBA can be 470 obtained and are listed in Fig.18 which shows that the PVC tube confinement did not provide 471 enhancement in the compressive strength of the RAC-RCBA, which may be attributed to the 472 very low confining pressure provided by the PVC tube, as listed in Eq. (1), the lateral 473 confining pressure is high dependent on the tensile strength and the thickness of the PVC, 474 both of which are relatively small. The 6-layer PFRP tube confinement resulted in a 26% 475 increase in the compressive strength of the RAC-RCBA. Overall, both PFRP tube-PVC and 476 PFRP strip-PVC confinement enhanced the compressive strength of RAC-RCBA effectively. 477 Comparison of PFRP tube-PVC-RAC-RCBA with different thickness of PFRP tube (i.e., 3P-478 T, 6P-T and 9P-T) demonstrated that with an increase of number of the PFRP layers, the 479 increment of compressive strength of the RAC-RCBA enhanced. For PFRP strip-PVC-RAC-480 RCBA specimens with the same strip thickness but different strip spacing (i.e., 3P-S25 and 481 3P-S50, 6P-S25 and 6P-S50, 9P-S25 and 9P-S50), increasing the spacing distance of the 482 PFRP strip reduced the increments to some extent. Besides, the PFRP tube-PVC-RAC-RCBA 483 had larger compressive strength than the corresponding PFRP strip-PVC-RAC-RCBA with 484 the same PFRP and PVC thickness. 485 486 The ductility indices µ of all the confined specimens in Fig. 19 were calculated as the ratio of 487

the cu to the co . The PVC-RAC-RCBA showed an obvious improvement in the ductility, 488

which was 4.2. The 6-layer PFRP-RAC-RCBA also had a significantly enhanced ductility 489 index, which was 13.4. Comparison of PFRP tube-PVC-RAC-RCBA (i.e., 3P-T, 6P-T and 490 9P-T) with different PFRP tube thicknesses demonstrated that with an increase of number of 491 the PFRP layers, the ductility index also increased remarkably. In general, the ductility index 492 of the PFRP tube-PVC-RAC-RCBA specimen was larger than that of the corresponding 493

19

PFRP strip-PVC-RAC-RCBA with the same thickness of PFRP. For the PFRP strip-PVC-494 RAC-RCBA specimens, an increase in the PFRP strip thickness also increased the ductility 495 index remarkably and an increase in the strip spacing reduced the inductility index of the 496 specimen. It should be pointed out here that the comparison in strength and ductility index is 497 based on the limited number of replications and the percentage increments are based on the 498 comparison on average values of tested group. 499

Figure 18: Increment ratios of compressive strength of different confined

RAC-RCBA specimens

Figure 19. Ductility indexes of the confined specimens

3.3.2 Comparison with GFRP-RAC-RCBA and CFRP-RAC-RCBA 500

In this section, the results of current study were compared with those of GFRP tube encased 501 and CFRP tube encased RAC-RCBA specimens in Ref. [21]. The compressive strength of the 502

21.7

8.6

4.5

34.2

28.7

17.8

39.6

30.5

19.8

26.0

0.1

3P-T 3P-S25 3P-S50 6P-T 6P-S25 6P-S50 9P-T 9P-S25 9P-S50 PFRP-RAC PVC-RAC

0

10

20

30

40

50

Specimens

Incr

emen

t o

f co

mp

ress

ive

stre

ng

th

(%)

8.6

6.6

5.3

13.4

9.2

6.3

15.0

10.7

8.3

13.3

4.2

21.7

8.6

4.5

34.2

28.7

17.8

39.6

30.5

19.8

26.0

0.1

3P-T 3P-S25 3P-S50 6P-T 6P-S25 6P-S50 9P-T 9P-S25 9P-S50 PFRP-RACPVC-RAC0

2

4

6

8

10

12

14

16

Duct

ilit

y i

nd

ex u

Specimens

20

unconfined RAC-RCBA, the replacement ratio of RAs for the RAC-RCBA, the constitutive 503 of the RAs, and the size of the RAC-RCBA used in Ref. [21] were similar as those used in 504 current study. The comparison in results of 6-layer CFRP-RAC-RCBA, 6-layer GFRP-RAC-505 RCBA, 6-layer PFRP-RAC-RCBA and 6-layer PFRP tube-PVC-RAC-RCBA is listed in 506 Table 7. The comparison in the tensile properties of PFRP, PVC, GFRP and CFRP can be 507 found in Table 5. 508 509 The comparison indicated that the 6-layer CFRP or 6-layer GFRP tube encased RAC-RCBA 510 had much higher compressive strength than that of 6-layer PFRP-RAC-RCBA or 6-layer 511 PFRP tube-PVC-RAC-RCBA, while the ductility indices of 6-layer CFRP-RAC-RCBA and 512 6-layer GFRP-RAC-RCBA were much lower than that of the 6-layer PFRP-RAC-RCBA or 513 6-layer PFRP-PVC-RAC-RCBA. The confinement ratio is defined as fcu/fco to evaluate the 514 effectiveness of FRP confinement on concrete, and the ratios of confinement ratio and cost 515 were calculated to compare the cost performance as Table 7. Although the confinement ratios 516 of PFRP-RAC-RCBA and PFRP tube-RAC-RCBA are lower than those of CFRP-RAC-517 RCBA and GFRP-RAC-RCBA, the ratios of confinement ratio and cost of PFRP-RAC-518 RCBA and PFRP tube-PVC-RAC-RCBA are much higher than those of CFRP-RAC-RCBA 519 and GFRP-RAC-RCBA. 520 521

Table 7 Comparison of CFRP, GFRP and PFRP tube confined RAC-RCBA 522

Specimens fco

(MPa)

ɛco

(%) fcu

(MPa)

ɛcu

(%) 𝜇

fl

(MPa) fl/fco fcu/fco

Price of

fibers

(RMB/m2)

Confinement

ratio/Cost

6L-CFRP-

RAC-RCBA 33.3 0.3 161.3 1.71 5.6 40.7 1.22 4.84

Glass fibre

80 0.61

6L GFRP-

RAC-RCBA 33.3 0.3 148.6 2.28 7.7 33.3 1 4.46

Carbon fibre

160 0.07

6L PFRP-

PVC-RAC-

RCBA

30.5 0.2 40.9 2.54 13.4 3.7 0.12 1.34 Polyester

fibre 1.1 1.22

6L PFRP-

RAC-RCBA 30.5 0.2 38.4 2.55 13.3 3.2 0.10 1.26

Polyester

fibre 1.1 1.14

523

3.4 Dilation behavior 524 In this section, the discussion on the dilation behavior of PFRP-PVC-RAC-RCBA specimens 525 is presented. The dilation effect was caused by the expansion of the core concrete when the 526 applied stress was close to or exceeded the ultimate axial stress of the concrete and then the 527 outer confining tubes were activated to dilate. The dilation rate 𝜇𝑡 is expressed as the slope of 528 the lateral strain increment to axial strain increment and given as Eq. (2) [67]. The 529 corresponding dilation rates of PFRP-PVC-RAC-RCBA are presented in Fig.20. 530

𝜇𝑡 = ∆𝜀h/∆𝜀𝑐 (2) 531 As illustrated in Fig.20, the peak dilation rate of PVC-RAC-RCBA is considerable indicating 532 the significant improvement in ductility due to the PVC confinement (as illustrated in Fig 19). 533 The peak dilation rate of the 6-layer PFRP-RAC-RCBA is like that of 6-layer PFPR tube-534 RAC-RCBA. For PFRP tube-PVC-RAC-RCBA with different PFRP thicknesses (i.e., 3P, 6P 535 and 9P), both the ultimate axial strains and the peak dilation rates increase with an increase of 536 the PFRP thickness. For PFPR strip-PVC-RAC-RCBA with different strip spacing (i.e., 3P-537 S25 and 3P-S50, 6P-S25 and 6P-S50, 9P-S25 and 9P-S50), a decrease of spacing distance of 538 the PFRP strip enhances the peak dilation rate with more ductile characteristic of specimens. 539 In addition, the PFRP tube-PVC-RAC-RCBA specimens always present larger dilation rate 540 than the corresponding PFRP strip-PVC-RAC-RCBA specimens with the same PFRP 541 thickness. 542

21

Fig.20 Dilation rates of PFRP-PVC-RAC-RCBA specimens

3.5 Axial strain at mid height and ends of the tubes 543 The average axial strains at the mid height and ends of the PFRP tube-PVC-RAC-RCBA 544 specimens (i.e., 3P-T 6P-T and 9P-T specimens) were obtained from the four strain gauges at 545 the mid height on the surface of the PFRP tube (i.e., SG5, SG6, SG7 and SG8, see Fig 12) and 546 four axial strain gauges at both the ends at the surface of the PFRP tube (i.e., SG9, SG10, 547 SG11 and SG12, see Fig 12). The comparisons of the stress and axial strain at the mid height 548 and both ends of specimens (i.e., 3P-T, 6P-T and 9P-T) are shown in Fig. 21. Generally, the 549 overall tendency of the stress-strain behavior at the middle region of specimens is similar to 550 that at the ends of specimens. Apparently, the axial deformation at the middle region of 551 specimens is larger than that at the ends of specimens under the same stress. The possible 552 reasons include: (1) the larger expansion of core concrete at the middle region under the 553 compression, and (2) the existing of two extra PFRP strip strengthening at both the ends of 554 specimens in order to avoid premature rupture of ends of specimens. The difference in the 555 axial strain between the mid-height and ends of specimens was more pronouncedly for 556 specimens with less thick of the PFRP tube. 557

0 0.005 0.01 0.015 0.02 0.025 0.03 0.0350

0.5

1

1.5

2

Dil

atio

n r

ate

Axial strain

9-layer PFRP tube-PVC-RAC-RCBA

9-layer and 25 mm PFRP strip-PVC-RAC-RCBA

9-layer and 50 mm PFRP strip-PVC-RAC-RCBA

6-layer PFRP tube-PVC-RAC-RCBA

6-layer and 25 mm PFRP strip-PVC-RAC-RCBA

6-layer and 50 mm PFRP strip-PVC-RAC-RCBA

3-layer PFRP tube-PVC-RAC-RCBA

3-layer and 25 mm PFRP strip-PVC-RAC-RCBA

3-layer and 50 mm PFRP strip-PVC-RAC-RCBA

6-layer PFRP-RAC

4 mm PVC-RAC

22

a) 3-layer PFRP

b) 6-layer PFRP

0

5

10

15

20

25

30

35

40

45

50

0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016

3P-T

Axial Strain

Str

ess(

MP

a)

Ends

Middle

0

5

10

15

20

25

30

35

40

45

50

0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016

6P-T

Axial Strain

Str

ess(

MP

a)

Ends

Middle

23

c) 9-layer PFRP

Fig.21 End and middle stress-strain curves of PFRP tube-PVC- RAC-RCBA cylinders

4 Analytical modeling of PFRP-PVC-RAC-RCBA 558 4.1 Mechanical characteristics analysis of PFRP-PVC-RAC-RCBA in compression 559 For the purpose of designing PFRP-PVC-RCBA-RCBA structures for practical application, 560 accurate stress-strain models should be developed. To develop stress-strain models, it is 561 important for us to understand the mechanical characteristics of PFRP-PVC-RAC-RCBA in 562 axial compression. In this study, the mechanical characteristics analysis of PFRP-PVC-RAC-563 RCBA in axial compression followed the mechanical analysis of FRP-steel tube confined 564 concrete introduced by Teng et al. [60]. With the similar analysis, the mechanical 565 characteristics of PFRP-PVC-RAC-RCBA in axial compression can be obtained and is 566 illustrated in Fig.22. The core concrete is under a tri-axial compression state, therefore the 567 load carrying capacity and deformation capacity of the concrete are enhanced. From Fig. 22, 568 it is clear that both PFRP and the PVC provided the lateral confining pressure to the concrete 569 core, as expressed by Eq. (3): 570

𝑓𝑙′ = 𝑓𝑙𝑓 + 𝑓𝑙𝑝 (3) 571

As shown in Fig 22(b), the force equilibrium in the PVC tube can be expressed by Eq. (4): 572

𝑓𝑙𝑝 =2𝐸𝑃𝑉𝐶𝜀𝑃𝑉𝐶𝑡𝑃𝑉𝐶

𝑑 (4) 573

As shown in Fig 22(a), the force equilibrium in the PFRP section can be expressed by Eq. (5): 574

𝑓𝑙𝑓 =2𝐸𝑓𝑟𝑝𝜀𝑓𝑟𝑝𝑡𝑓𝑟𝑝

𝑑+2𝑡𝑃𝑉𝐶 (5) 575

Where fl’ is the lateral confining pressure of the composite confinement on concrete core, flf is 576

the effective lateral confining pressure provided by PFRP, flp is lateral confining pressure 577 provided by PVC, Epvc, ɛpvc, and tpvc are the elastic modulus, tensile strain in the hoop direction 578 and thickness of the PVC tube, Efrp, ɛfrp, and tfrp are the elastic modulus, tensile strain in the 579 hoop direction and thickness of PFRP tube, d is the diameter of the core concrete. 580 For PFRP strip-PVC-RAC-RCBA cylinders, the equivalent PFRP thickness tfrp

’ can be written 581 by Eq.(6): 582

𝑡𝑓𝑟𝑝′ =

𝑛𝑏𝑓𝑟𝑝

𝐻𝑡𝑓𝑟𝑝 (6) 583

Then, the force equilibrium of the PFRP strip confinement can be written as following Eq.(7):: 584

𝑓𝑙𝑓 =2𝑛𝐸𝑓𝑟𝑝𝜀𝑓𝑟𝑝𝑏𝑓𝑟𝑝𝑡𝑓𝑟𝑝

𝐻(𝑑+2𝑡𝑃𝑉𝐶) (7) 585

Where n is the number of PFRP strips, bfrp is the width of PFRP strip, H is the height of 586 concrete cylinders. The lateral confining pressure provided by the PFRP strips was imposed 587 onto the concrete core through the PVC to, the dispersion angle was 45° and the dispersion 588

0

5

10

15

20

25

30

35

40

45

50

0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016

Axial strain

Str

ess

(MP

a)

Ends

Middle

9P-T

24

width was the thickness of the PVC tube (Fig 23), as explained by Teng et al. [65]. Thus, the 589 effective lateral confining pressure onto the concrete core can be expressed by 590

𝑓𝑙𝑓′ = 𝑘𝑒𝑓𝑙𝑓 (8) 591

𝑘𝑒 =𝐴𝑒

𝐴 (9) 592

Ae = A−m(s−2tpvc)

2

2 (10) 593

𝐴 = 𝑑𝐻 (11) 594 Where ke is the effective confining coefficient of the PFRP strip, and ke =1 is for PFRP tube 595 confinement, Ae is the effective confining area of the core concrete, A is the gross area of 596 specimens including the core concrete and external hybrid tube, m is the number of zone 597 without confinement among PFRP strip, s is the spacing distance of the PFRP strip. Overall, 598 the Eq.(3) can be expressed as Eq.(12): 599

𝑓𝑙′ = 𝑓𝑙𝑓

′ + 𝑓𝑙𝑝 = 𝑘𝑒𝑓𝑙𝑓 + 𝑓𝑙𝑝 (12) 600

Specifically, for PFRP tube-PVC-RAC-RCBA cylinders: 601

𝑓𝑙′ =

2𝐸𝑓𝑟𝑝𝜀𝑓𝑟𝑝𝑡𝑓𝑟𝑝

𝑑+2𝑡𝑃𝑉𝐶+

2𝐸𝑃𝑉𝐶𝜀𝑃𝑉𝐶𝑡𝑃𝑉𝐶

𝑑 (13) 602

For PFRP strip-PVC-RAC-RCBA cylinders: 603

𝑓𝑙′ = 𝑘𝑒

2𝑛𝐸𝑓𝑟𝑝𝜀𝑓𝑟𝑝𝑡𝑓𝑟𝑝

𝐻(𝑑+2𝑡𝑃𝑉𝐶)+

2𝐸𝑃𝑉𝐶𝜀𝑃𝑉𝐶𝑡𝑃𝑉𝐶

𝑑 (14) 604

605

a) Force equilibrium of PFRP b) Force equilibrium of PVC c) Force equilibrium of RAC-RCBA

Fig.22 Mechanical characteristics of PFRP tube-PVC-RAC-RCBA cylinder

606

Fig.23 Mechanical characteristics of PFRP strip-PVC- RAC-RCBA cylinder

4.2 Strength and corresponding strain models for PFRP-PVC-RAC-RCBA specimens 607 The discussion in Section 3.3 shows that the RAC-RCBA cylinders were weakly-confined by 608 the PFRP-PVC tube although remarkably enhancement in compressive strength and ductility 609 were achieved. That is to say, the strain of PFRP-PVC-RAC-RCBA at peak stress at TP point 610 in Fig.15 (i.e., compressive strength) point named peak strain was not the ultimate strain of 611 PFRP-PVC-RAC-RCBA (i.e., UP point in Fig.15) where the hybrid confined system failed. 612 Generally, the ultimate stress and ultimate strain were related to the elastic modulus of 613 confined materials and core concrete [62], in order to fit the strength model, the eigenvalue λ 614

d

pvc

t

pvc pvc pvcE tpvc pvc pvcE t

frp

t

frp frp frpE t

flf f pl lf f

f pl lf f

flf

frp frp frpE t

45°

PVC

PFRP

s

Core concrete

A A

A - A

Core concrete

Zone without

confinement

s

s-2tpvc

bp

tpvc Effective confined zone

25

was cited as Eq.(15) [62] for better expressing the relationship of ultimate stress and the 615 elastic modulus of confined materials and core concrete: 616

𝜆 =𝐸𝑙

𝐸𝑐 (15) 617

Where Ec is the elastic modulus of the RAC-RCBA which is related to the square root of 618

compressive strength √𝑓𝑐𝑜 [61, 62] as Eq.(16), El is the effective lateral confining stiffness of 619

the hybrid PFRP-PVC tube which is expressed as Eq.(17) and Eq.(18): 620

𝜆 =𝐸𝑙

√𝑓𝑐𝑜 (16) 621

𝐸𝑙 =2𝐸𝑓𝑟𝑝𝑡𝑓𝑟𝑝

𝑑+2𝑡𝑃𝑉𝐶+

2𝐸𝑃𝑉𝐶𝑡𝑃𝑉𝐶

𝑑 (For PFRP tube-PVC-RAC-RCBA) (17) 622

𝐸𝑙 = 𝑘𝑒2𝑛𝐸𝑓𝑟𝑝𝑡𝑓𝑟𝑝

𝐻(𝑑+2𝑡𝑃𝑉𝐶)+

2𝐸𝑃𝑉𝐶𝑡𝑃𝑉𝐶

𝑑 (For PFRP strip-PVC-RAC-RCBA) (18) 623

The strength fcc’ and corresponding strain εcc

’ could be calculated as: 624 𝑓𝑐𝑐′

𝑓𝑐𝑜= 1 + 𝑘1(

𝜆

𝑓𝑐𝑜)𝛼 (19) 625

𝜀𝑐𝑐′

𝜀𝑐𝑜= 1 + 𝑘2(

𝜆

𝜀𝑐𝑜)𝛽 (20) 626

627 Based on the regression analysis and iterative computations of the tested results, the fitting 628 coefficients α=0.622 and k1=0.405, β=0.684 and k2=1.647 can be determined to create the 629 strength and strain equations for PFRP-PVC-RAC-RCBA, as illustrated in Fig 24. The 630 strength and peak strain models for PFRP-PVC-RAC-RCBA can be expressed by Eq.(21) and 631 (22): 632

𝑓𝑐𝑐′

𝑓𝑐𝑜= 1 + 0.405(

𝜆

𝑓𝑐𝑜)0.622 (21) 633

𝜀𝑐𝑐′

𝜀𝑐𝑜= 1 + 1.647(

𝜆

𝜀𝑐𝑜)0.684 (22) 634

0 0.2 0.4 0.6 0.8 1 1.2 1.40.8

1.0

1.2

1.4

1.6

622.0

''

'

405.01

coco

cc

ff

f

λ / f 'co

f

'cc

/ f ' co

R2=0.78

a) Fitting curves of strength model

26

0 0.2 0.4 0.6 0.8 1 1.2 1.40.5

1.0

1.5

2.0

2.5

3.0

3.5

684.0

'

cc 647.11

coco f

ε c

c /

εco

λ / f 'co

R2=0.85

b) Fitting curves of peak strain model

Fig. 24 Fitting curves of peak strength models

Based on the stress and strain equations above, the comparison in the peak strength and the

peak axial strain between the experimental and predictions is shown in Fig 25. It can be seen

that the predictions based on the developed stress and strain models matched the experimental

stress and strain values of the PFRP-PVC-RAC-RCBA well, with relatively small deviations.

a) Performance of strength model

10 15 20 25 30 35 40 45 5010

15

20

25

30

35

40

45

50

Calculated value/Tested value

AVG=1.005

SD=0.046

Cal

cula

ted v

alue

(MP

a)

Tested value (MPa)

27

b) Performance of peak strain model

Fig. 25 Performance of peak strength models: strength and peak strain models

4.3 Ultimate strain and corresponding stress models 635 In literature, the ultimate strain and corresponding stress models for FRP confined concrete 636 typically had the following expressions [63, 64]: 637

𝑓𝑐𝑢′

𝑓𝑐𝑜= 𝑐1 + 𝑘3(

𝑓𝑙

𝑓𝑐𝑜)𝑏1 (23) 638

𝜀𝑐𝑢′

𝜀𝑐𝑜= 𝑐2 + 𝑘4(

𝑓𝑙

𝑓𝑐𝑜)𝑏2 (24) 639

Where 𝑓𝑐𝑢′ and

'

cu are the ultimate stress and the corresponding ultimate strain of PFRP-640

PVC-RAC-RCBA as UP point in Fig.15. Based on the regression analysis and iterative 641 computations of the partial tested results, the fitting coefficients c1=1.17, b1=-1.665 and k3=-642 0.0043, c2=1, b2=0.817 and k4=63.64 can be determined to create the strain equations for 643 PFRP-PVC-RAC-RCBA, as illustrated in Fig.26. The strength and peak strain models can be 644 expressed by Eq.(25) and (26): 645

646 𝑓𝑐𝑢′

𝑓𝑐𝑜= 1.17 − 0.0043(

𝑓𝑙

𝑓𝑐𝑜)−1.665 (25) 647

𝜀𝑐𝑢′

𝜀𝑐𝑜= 1 + 63.64(

𝑓𝑙

𝑓𝑐𝑜)0.817 (26) 648

0.1 0.2 0.3 0.4 0.5 0.60.1

0.2

0.3

0.4

0.5

0.6

Cal

cula

ted v

alue

(%)

Tested value (%)

Calculated value/Tested value

AVG=1.027

SD=0.131

28

a) Fitting curves of ultimate stress model

b) Fitting curves of ultimate strain model

Fig.26 Fitting curves of ultimate models

a) Fitting curves of ultimate stress model

0.00 0.04 0.08 0.12 0.16 0.200.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.665cu

co co

1.17 0.0043( )'

l

' '

f f

f f

f l / f

co

f ' cu

/ f co

R2=0.81

模型 akc (User)

方程 a+b*x^c

绘图 B

a 1.17059 ± 0.10869

b -0.0043 ± 0.00803

c -1.66522 ± 0.58365

Reduced Chi-Sqr 0.00894

R平方(COD) 0.81359

调整后R平方 0.79805

0.00 0.04 0.08 0.12 0.16 0.200

2

4

6

8

10

12

14

16

18

0.817cu

co co

1+63.64( )l

'

f

f

f l / f

co

ε cu

/ ε

co

R2=0.94

10 15 20 25 30 35 40 45 5010

15

20

25

30

35

40

45

50

Cal

cula

ted

val

ue

(MP

a)

Tested value (MPa)

Calculated value/Tested value

AVG=1.013

SD=0.115

29

b) Fitting curves of ultimate strain model

Fig.27 Fitting curves of ultimate models 649 Based on the ultimate stress and strain equations above, the comparison in the ultimate stress 650 and the ultimate axial strain between the experimental and predictions is shown in Fig 27. It 651 can be seen that the predictions based on the developed stress and strain models matched the 652 experimental stress and strain values of the PFRP-PVC-RAC-RCBA well, with relatively 653 small deviations. 654 655 5 Conclusions 656 This paper investigated the axial compression behavior of polyester FRP-PVC tube encased 657 recycled aggregate concrete (RAC) with recycled clay brick aggregates (RCBA). Two 658 experimental phases were carried out. In the first stage, different replacement ratios of 659 recycled aggregates and different cement-water ratios for RAC-RCBAs were considered in 660 order to find out the optimized relationship between the strength of RAC and replacement 661 ratios and cement-water ratios. In the second stage, experimental works were carried out to 662 investigate the effects of type of PFRP confinement (i.e. tube and strip), PFRP thickness and 663 spacing of PFRP strips on the axial compressive behavior of PFRP-PVC-RAC-RCBA 664 specimens. The study reveals that: 665 1. The compressive strength of RAC-RCBA decreased with an increase of the water-cement 666

ratios. The RAC-RCBA with 70% replacement ratio of RAs obtained the highest 667 compressive strength. In general, the compressive strengths of all the RAC-RCBAs were 668 lower than that of the NAC. The failure mode of RAC-RCBA was similar to that of NAC. 669 However, the close-up showed that most of the RCBA coarse aggregates in the RAC 670 were broken under compression but this was not observed for the natural coarse 671 aggregates in the NAC. This is attributed to the much larger crushing of the RAs with 672 RCBA than that of the NAs for NAC. 673

2. The PFRP tube, PFRP-strip-PVC tube and PFRP-tube-PVC tube confinement all 674 increased the load carrying capacity and ductility of the RAC-RCBA remarkably. The 675 PVC tube confinement did not show enhancement in the compressive strength but 676 resulted in significant enhancement in the ductility. In general, the enhancement in 677 compressive strength and ductility by PFRP-PVC tube dual confinement was larger than 678 those of the corresponding PFRP or PVC tube single confinement, i.e. the increment in 679 strength of RAC-RCBA by the 6-layer PFRP tube-PVC confinement was 34.2%, while 680 that of RAC-RCBA by 6-layer PFRP or PVC tube alone was 26.0% and 0.1%, 681 respectively. 682

3. The improvement on strength and ductility for the RAC-RCBA by the PFRP-PVC 683 confinement was increase with an increasing PFRP tube thickness and a decreasing 684

0.00 0.01 0.02 0.03 0.04 0.05 0.060.00

0.01

0.02

0.03

0.04

0.05

0.06

Cal

cula

ted

val

ue

Tested value

Calculated value/Tested value

AVG=1.020

SD=0.092

30

spacing of the PFRP strip. In addition, the PFRP tube-PVC-RAC-RCBA always exhibited 685 higher compressive strength and ultimate strain than those of the corresponding PFRP 686 strip-PVC-RAC-RCBA with the same PFRP thickness. However, the PFRP strip-PVC-687 RAC-RCBA showed the same compressive stress-strain curve pattern with the PFRP 688 tube-PVC-RAC-RCBA. 689

4. The comparison with the GFRP tube RAC-RCBA and CFRP tube RAC-RCBA showed 690 that GFRP and CFRP confinement resulted in much higher enhancement in the 691 compressive strength compared with PFRP, PVC and PFRP-PVC for RAC-RCBA due to 692 the much larger tensile strength and modulus of the GFRP and CFRP. However, the 693 PFRP and PFRP-PCV confinement resulted in much larger ductility compared with their 694 GFRP and CFRP confinement counterparts. 695

5. Design-oriented stress-strain models were developed for PFRP-PVC-RAC-RCBA 696 specimens in axial compression. The accuracy of the developed models was verified with 697 the experimental results. 698

Overall, this study confirmed that the PFRP-PVC-RAC-RCBA hybrid system is quite 699 promising for structural application with desirable carrying capability and ductility 700 characteristic. In the future study, the effects of different experimental parameters such as size 701 effect and slenderness ratio of the specimens on axial compression and even dynamic loading 702 responses need to be evaluated. 703 704

Acknowledgement 705 This study is supported by National Key Research and Development Program of China (No. 706 2017YFC0703305) and Federal Ministry of Education and Research of Germany (No. 707 01DS18023). 708 709 References 710

[1] Ghisellini P, Ji X, Liu GY, Ulgiati S. Evaluating the transition towards cleaner production 711 in the construction and demolition sector of China: A review. Journal of Cleaner 712 Production 2018;195: 418-434 713

[2] Nezerka V, Hrbek V, Prosek Z, Somr M. Micromechanical characterization and 714 modelling of cement pastes containing waste marble power. Journal of Cleaner 715 Production 2018;195: 1081-1090. 716

[3] J.L. Chen. Reconsideration of issues about construction waste recycling. Construction 717 Technology. 2015;58-59. 718

[4] J.L. Chen, W.J. Zhou, W. Li. Precondition and problems in reutilization of building 719 wastes. Architecture Technology. 2015;46(12):1114-1116. 720

[5] T.U. Mohammed, A. Hasnat, M.A Awal. Recycling of brick aggregate concrete as coarse 721 aggregate. Journal of Materials in Civil Engineering. 2014;27(7): B4014005. 722

[6] J. Yang, Q. Du, Y.W. Bao. Concrete with recycled concrete aggregate and crushed clay 723 bricks. Construction and Building Materials. 2011;25:1935-1945. 724

[7] C.S Poon., Z.H. Shui, L Lam. Effect of microstructure of ITZ on compressive strength of 725 concrete prepared with recycled aggregates. Construction and Building Materials. 2004; 726 18(6): 461-468. 727

[8] M. Etxeberria, E. Vázquez, A. Marí. Influence of amount of recycled coarse aggregates 728 and production process on properties of recycled aggregate concrete. Cement and 729 concrete research. 2007;37(5):735-742. 730

[9] Akhtat A, Sarmah A. Strength improvement of recycled aggregate concrete through 731 silicon rich char derived from organic waste. Journal of Cleaner Production 2018;196: 732 411-423. 733

[10] X. Sun, L. Yan, Kasal B. Impact behavior of concrete columns confined by both GFRP 734 tube and steel spiral reinforcement. Construction and Building Materials. 2017;131:438-735 448. 736

31

[11] L. Yan, N. Chouw, K. Jayaraman. Effect of column parameters on flax FRP confined 737 coir fibre reinforced concrete. Construction and Building Materials. 2014;55:299-312. 738

[12] L. Yan, N, Chouw. Natural FRP tube confined fibre reinforced concrete under pure axial 739 compression: A comparison with glass/carbon FRP. Thin-Walled Structures. 740 2014;82:159-169. 741

[13] G. Ma, H. Li. Testing and analysis of basalt FRP-confined damaged concrete cylinders 742 under axial compression loading. Construction and Building Materials. 2018;169:762-743 774. 744

[14] L. Yan, N. Chouw. Experimental study of flax FRP tube encased coir fibre reinforced 745 composite column. Construction and Building Materials. 2013;40:1118-1127. 746

[15] A. Duchez. Effect of bond on compressive behavior of flax fibre reinforced polymer 747 tube-confined coir fibre reinforced concrete. Journal of Reinforced Plastics and 748 Composites. 2013;32:273-285. 749

[16] L. Huang, P. Yin, L. Yan, B. Kasal. Behavior of hybrid GFRP-perforated-steel tube 750 encased concrete under uniaxial compression. Composite Structures. 2016;142:313-324. 751

[17] C. Gao, L. Huang, L. Yan, G. Ma, L. Xu, Compressive behavior of CFFT with inner 752 steel wire mesh, Composite Structures. 2015;133:322-330. 753

[18] L. Huang, X. Xun, D. Zhu. Compressive behavior of concrete confined with GFRP 754 tubes and steel spirals. Polymers. 2015;7 (5):851-875. 755

[19] L. Huang, D. Zhu. Compressive behavior of concrete confined by CFRP and transverse 756 spiral reinforcement. Part A: experimental study. Material Structures. 2016;49(3):1001-757 1011. 758

[20] J. Xiao, Y. Huang, J. Yang. Mechanical properties of confined recycled aggregate 759 concrete under axial compression. Construction and Building Materials. 2012;26(1): 591-760 603. 761

[21] C. Gao, L. Huang, L. Yan, B. Kasal, W. Li. Behavior of glass and carbon FRP tube 762 encased recycled aggregate concrete with recycled clay brick aggregate. Composite 763 Structures. 2016;155:245-254. 764

[22] T. Xie, T. Ozbakkaloglu. Behavior of recycled aggregate concrete-filled basalt and 765 carbon FRP tubes. Construction and Building Materials. 2016;105:132-143. 766

[23] G.M. Chen, Y.H. He, T. Jiang. Behavior of CFRP-confined recycled aggregate concrete 767 under axial compression. Construction and Building Materials. 2016;111:85-97. 768

[24] M.S.I. Choudhury, A.F.M.S. Amin, M.M. Islam, A. Hasnat. Effect of confining pressure 769 distribution on the dilation behavior in FRP-confined plain concrete columns using stone, 770 brick and recycled aggregates. Construction and Building Materials. 2016:102(1):541-771 551. 772

[25] Ardavan Yazdanbakhsh, Lawrence C.B. The effect of shear strength on load capacity of 773 FRP strengthened beams with recycled concrete aggregate. Construction and Building 774 Materials. 2016:102(1):133-140. 775

[26] R.D. Freeman, J.L. Burati, S.N. Amirkhanian. Polyester fibers in asphalt paving 776 mixtures. Association of Asphalt Paving Technologists Proc. 1989;58. 777

[27] Z. Jing, Z. Bayasi. Properties of polyester fiber reinforced concrete. Journal of Dalian 778 University of Technology. 1993;S2. 779

[28] T. Ochi, S. Okubo, K. Fukui. Development of recycled PET fiber and its application as 780 concrete-reinforcing fiber. Cement and Concrete Composites. 2007;29(6):448-455. 781

[29] J.H.J. Kim, C.G. Park, S.W. Lee. Effects of the geometry of recycled PET fiber 782 reinforcement on shrinkage cracking of cement-based composites. Composites Part B: 783 Engineering. 2008;39(3): 442-450. 784

[30] De Oliveira L.A.P, Castro-Gomes J.P. Physical and mechanical behavior of recycled 785 PET fibre reinforced mortar. Construction and Building Materials. 2011;25(4):1712-1717. 786

[31] R.P. Borg, O. Baldacchino, L. Ferrara. Early age performance and mechanical 787 characteristics of recycled PET fibre reinforced concrete. Construction and Building 788 Materials. 2016;108:29-47. 789

[32] J.G. Dai, Y.L. Bai, J.G. Teng. Behavior and Modeling of Concrete Confined with FRP 790

32

Composites of Large Deformability. Journal of Composites for Construction. 2011; 791 15(6):963-973. 792

[33] Ispir M. Monotonic and cyclic compression tests on concrete confined with PET-FRP. 793 Journal of Composites for Construction. 2014;19(1): 04014034. 794

[34] S. Saleem, Q. Hussain, A. Pimanmas. Compressive behavior of PET FRP-confined 795 circular, square, and rectangular concrete columns. Journal of Composites for 796 Construction. 2016;21(3): 04016097. 797

[35] S. Saleem, A. Pimanmas, W. Rattanapitikon. Lateral response of PET FRP-confined 798 concrete. Construction and Building Materials. 2018;159: 390-407. 799

[36] A. Pimanmas, S. Saleem. Dilation Characteristics of PET FRP-Confined Concrete. 800 Journal of Composites for Construction. 2018;22(3): 04018006. 801

[37] L. Huang, X. Yang, L. Yan, K. He, H. Li, Y. Du. Experimental study of polyester fiber-802 reinforced polymer confined concrete cylinders. Textile Research Journal. 2016; 803 86(15):1606-1615. 804

[38] L. Huang, L. Chen, L. Yan. Behavior of polyester FRP tube encased recycled aggregate 805 concrete with recycled clay brick aggregate: Size and slenderness ratio effects. 806 Construction and Building Materials. 2017;154:123-136. 807

[39] T.A. Ranney, L.V. Parker. Susceptibility of ABS, FEP, FRE, FRP, PTFE, and PVC well 808 casings to degradation by chemicals. Cold Regions Research and Engineering Lab 809 Hanover NH. 1995. 810

[40] Al Malaika S., Golovoy A., Wilkie C.A. Chemistry and technology of polymer additives. 811 Blackwell Science. 1999. 812

[41] Awham M.H., Salih Z.G.M. A study of some mechanical behavior on a thermoplastic 813 material. Journal of Al-Nahrain University. 2011;14(3):58-65. 814

[42] Rohe F.P.S. Strength testing of thermal welded PVC geomembrane field seams using 815 non-destructive air channel methods. Environmental Protection, Inc. 816

[43] Nowack R., Otto O.I., Braun E.W. 60 jahre erfahrungen mit rohrleitungen aus 817 weichmacherfreiem polyvinylchlorid (PVC-U). KRV Nachrichten. 1995;1-95. 818

[44] Ranne T.A., Parker L.V. Susceptibility of ABS, FEP, FRE, FRP, PTFE, and PVC Well 819 Casing to Degradation by Chemical. Special Report 95-l, US Army Corps of 820 Engineerings, January 1995. 821

[45] Kurt C. E. Concrete filled structural plastic columns. Journal of the Structural Division, 822 1978, 104 (ASCE 13478 Proceeding). 823

[46] H. Toutanji. Behavior of concrete columns encased in PVC-FRP composite tubes. 824 Advanced Materials in Bridges and Structures (ACMBS-III). 2000;809-817. 825

[47] H. Toutanji, M. Saafi. Durability studies on concrete columns encased in PVC-FRP 826 composite tubes. Composite Structures. 2001;54(1): 27-35. 827

[48] Toutanji H., Saafi M. Stress-strain behavior of concrete columns confined with hybrid 828 composite materials. Materials & Structures, 2002, 35(6):338-347. 829

[49] Wang J., Yang Q. Experimental study on mechanical properties of concrete confined 830 with plastic pipe. ACI Mater J. 2010;107(2). 831

[50] Gupta P.K., Verma V.K. Study of concrete-filled unplasticized poly-vinyl chloride tubes 832 in marine environment. Proc Inst Mech Eng Part M J Eng Marit Environ. 833 2014;1475090214560448. 834

[51] Gathimba Naftary K., Oyawa Walter O., Mang’uriu Geoffrey N. Compressive strength 835 characteristics of concrete-filled plastic tubes short columns. Int J Sci Res (IJSR). 836 2014;3(9):2168-2174. 837

[52] Jiang S.F., Ma S.L., Wu Z.Q. Experimental study and theoretical analysis on slender 838 concrete-filled CFRP-PVC tubular columns. Construction and Building Materials. 839 2014;53:475-487. 840

[53] GB/T 25177-2010. Recycled coarse aggregate for concrete. Standardization 841 Administration of The People’s Republic of China. 2011. 842

[54] D. ASTM. Standard test method for tensile properties of polymer matrix composite 843 materials. 2008. 844

33

[55] GB/T 8804.2-2003. Thermoplastic pipes-Determination of tensile properties-Part 2: 845 Pipes made of unplasticized poly (vinyl chloride) (PVC-U), chloriniated poly (vinyl 846 chloride) (PVC-C) and high-impact poly (vinyl chloride) (PVC-HI). General 847 Administration of Quality Supervision, Inspection and Quarantine of the People's 848 Republic of China. 2003. 849

[56] Zhao L.J., Weng J.L., Chen W. Test and research on the double horizontal shaft 850 vibrating mixing technology. Construction Machinery Technology and Management. 851 2013;2:107-110. 852

[57] Feng X.N., Feng Z.X., Wang W.Z. Review on concrete vibratory mixing techniques. 853 Chinese Journal of Construction Machinery. 2007;5(1):113-116. 854

[58] T. Ozbakkaloglu. Compressive behavior of concrete-filled FRP tube columns: 855 Assessment of critical column parameters. Engineering Structures. 2013;51:188-199. 856

[59] M. Fakharifar, G. Chen. Compressive behavior of FRP-confined-filled PVC tubular 857 columns. Composite Structures. 2016;141:91-109. 858

[60] J.G. Teng, Y. Tao, W.Y. Long, S.L. Dong, Y.F. Yang. Behavior of hybrid FRP-859 concrete-steel tubular columns: experimental and theoretical studies. Progress in steel 860 building structures. 2006;8(5):1-7. 861

[61] Pantelides C.P., Yan Z. Confinement model of concrete with externally bonded FRP 862 jackets or post tensioned FRP shells. Journal of Structural Engineering. 863 2007;133(9):1288-1296. 864

[62] Xiao Y., Wu H. Compressive behavior of concrete confined by various types of FRP 865 composite jackets. Journal of Reinforced Plastics and Composites. 2003;22(13):1187-866 1201. 867

[63] L. Huang, C. Gao, L.B. Yan, B. Kasal, G. Ma. Reliability assessment of confinement 868 models of carbon fiber reinforced polymer-confined concrete. Journal of Reinforced 869 Plastics and Composites. 2016;0(0):1-31. 870

[64] L. Huang, C. Gao, L.B. Yan, B. Kasal, G. Ma, H.Z. Tan. Confinement models of GFPR-871 confined concrete: Statistical analysis and unified stress-strain models. Journal of 872 Reinforced Plastics and Composites. 2016;0(0):1-25. 873

[65] P.P. Li. The research of mechanical behavior of BFRP-PVC tube self-compacting 874 recycled concrete short column under uniaxial compression. Liaoning University of 875 Technology. 2015. 876

[66] Spoelstra M.R., Monti G. FRP-confined concrete model. Journal of Structural 877 Engineering. 1999;9(4):143-150. 878

[67] Jian C. Lim, T. Ozbakkaloglu. Lateral strain to axial strain relationship of confined 879 concrete. Engineering Structures. 2015;141(5):04014141. 880

[68] Riccio A., Di Costanzo C., Di Gennaro P., Sellitto A., Raimondo A. Intra-laminar 881 progressive failure analysis of composite laminates with a large notch damage. 882 Engineering Failure Analysis. 2017;73:97-112. 883

[69] Riccio A., Sellitto A., Saputo S., Russo A., Zarrelli M., Lopresto V. Modelling the 884 damage evolution in notched omega stiffened composite panels under compression. 885 Composites Part B-Engineering. 2017;126:60-71. 886

[70] Riccio A., Damiano M., Raimondo A., Di Felice G., Sellitto A. A fast numerical 887 procedure for the simulation of inter-laminar damage growth in stiffened composite 888 panels. Composite Structures. 2016;145:203-216. 889

[71] Campione G., La Mendola L., Monaco A., Valenza A., and Fiore V. Behavior in 890 compression of concrete cylinders externally wrapped with basalt fibers. Composites: 891 Part B. 2015;69:576-586. 892

[72] De Santis S., De Felice G., Napoli A., and Realfonzo R. Strengthening of structures with 893 Steel Reinforced Polymers: A state-of-the-art review. Composites: Part B. 2016;104:87-894 110. 895

896